CN108389315B

CN108389315B - Article identification method and apparatus, and computer-readable storage medium

Info

Publication number: CN108389315B
Application number: CN201810174634.7A
Authority: CN
Inventors: 张爱喜; 刘巍; 谭志羽; 陈宇; 翁志
Original assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Current assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Priority date: 2018-03-02
Filing date: 2018-03-02
Publication date: 2021-07-20
Anticipated expiration: 2038-03-02
Also published as: CN108389315A

Abstract

The invention discloses an article identification method and device and a computer readable storage medium, and relates to the field of data processing. The article identification method comprises the following steps: acquiring a weight change value in response to monitoring that the load bearing weight of the goods shelf changes; determining one or more article combinations corresponding to the weight change value from the unsettled articles according to the weight of each article; determining an article combination taken and placed by a user from one or more article combinations according to the attribute of each article combination; and updating the set of the articles to be settled according to the article combination taken and placed by the user. Therefore, the articles taken and placed by the user can be accurately determined, and the accuracy of article identification is improved. And the user can get goods, put goods, get goods and put goods simultaneously to increased the flexibility that the user used, promoted user experience. In addition, the invention has low manufacturing cost and is convenient to implement and deploy.

Description

Article identification method and apparatus, and computer-readable storage medium

Technical Field

The present invention relates to the field of data processing, and in particular, to an article identification method and apparatus, and a computer-readable storage medium.

Background

In the related art, the commodity Identification scheme of the vending machine is mainly an Identification scheme based on an RFID (Radio Frequency Identification) technology. In this scenario, an RFID tag needs to be provided for each item. When the consumer takes the goods, the goods pass through an area equipped with an RFID reader to be automatically sensed, thereby recognizing the goods and completing self-payment.

Another solution is to determine the type of goods based on image recognition. The condition of the container is shot by arranging the camera on the container, and interframe comparison detection is carried out according to certain frequency, so that whether the commodity area changes or not can be judged to identify the commodity purchased by a consumer.

However, the identification scheme in the related art has high cost and low identification rate.

Disclosure of Invention

After the inventor finds that, the technical scheme based on the RFID has the defect that the RFID tag is high in cost and not beneficial to scene replication and popularization for vending machine scenes with huge throughput. Meanwhile, when the RFID label meets liquid and metal, the attenuation and shielding are easy, the label is troublesome to paste and easy to tear, the size and the induction distance are difficult to coordinate, and the identification rate is reduced.

A disadvantage of image recognition-based solutions is that they are difficult to handle when several items, in particular several items of different types, are taken or returned at the same time. And for the extreme case of simultaneously taking and placing commodities, the image recognition is difficult to cover. And thus the recognition rate is also low.

The embodiment of the invention aims to solve the technical problem that: how to improve the accuracy of item identification in a vending scenario.

According to a first aspect of some embodiments of the present invention there is provided an item identification method comprising: acquiring a weight change value in response to monitoring that the load bearing weight of the goods shelf changes; determining one or more article combinations corresponding to the weight change value from the unsettled articles according to the weight of each article; determining an article combination taken and placed by a user from one or more article combinations according to the attribute of each article combination; and updating the set of the articles to be settled according to the article combination taken and placed by the user.

In some embodiments, each item combination comprises at least one item, the non-settlement items comprise a shelf item and an item to be settled, the shelf item is an item on the shelf before the bearing weight of the shelf changes, and the item to be settled is an item to be settled before the bearing weight of the shelf changes.

In some embodiments, determining, from the outstanding items, one or more item combinations to which the weight change value corresponds based on the weight of each item comprises: combining one or more of the outstanding items to obtain one or more item combinations; calculating the change weight corresponding to each article combination, wherein the change weight corresponding to each article combination is equal to the weight of the goods on the shelf in the article combination minus the weight of the goods to be settled, and the weight change value is the difference between the load weight of the shelf before and after the change; and determining the article combination with the difference between the changed weight and the weight change value within a preset range as the article combination corresponding to the weight change value.

In some embodiments, the corresponding varying weight of each combination of items is calculated from a pre-obtained average weight of each item; and determining the article combination with the weight deviation range including the weight change value as the article combination corresponding to the weight change value, wherein the weight deviation range is calculated according to the maximum weight deviation amount of each article in the combination.

In some embodiments, in response to monitoring that the load bearing weight of a unit weighing area on a shelf changes, obtaining a weight change value of the unit weighing area; and determining one or more article combinations corresponding to the weight change values from the unsettled articles according to the weight of each article, wherein the unsettled articles comprise articles to be settled and articles in the unit weighing area.

In some embodiments, a confidence level for each combination of items is determined based on the attributes of each combination of items; selecting the article combination with the highest confidence coefficient as the article combination taken and placed by the user; wherein the attribute of each combination of items comprises at least one of: the difference between the corresponding change weight of the article combination and the change value of the weight, whether the articles in the article combination simultaneously comprise the goods on the shelf and the articles to be settled, the article type related to the article combination, the number of the article types with the number of the articles being 1 in the article combination, and whether the articles in the article combination are the articles with the largest number in the current goods channel.

In some embodiments, the article identification method further comprises: and responding to the settlement of the article to be settled by the user, and emptying the article to be settled.

In some embodiments, the article identification method further comprises: acquiring the weight of each article of each to-be-selected category; performing at least one of the following screening operations: deleting the categories of which the weight is less than a preset value from the categories to be selected, and deleting at least one of two category combinations of which the weights have a multiple relation, wherein the category combination comprises one or more categories; and using the sorted goods as goods of the goods on the shelf.

In some embodiments, the article identification method further comprises: acquiring an article image shot when a user takes and places articles on a shelf; identifying the article image to obtain an identification result; determining a user-accessible combination of items from the one or more combinations of items based on the attributes of each combination of items comprises: and determining the article combination taken and placed by the user from one or more article combinations according to the attribute of each article combination and the identification result based on the image identification.

According to a second aspect of embodiments of the present invention, there is provided an article identification apparatus comprising: the weight change acquisition module is configured to respond to the monitored change of the bearing weight of the goods shelf and acquire a weight change value; an item combination determination module configured to determine one or more item combinations corresponding to weight variation values from the unsettled items according to the weight of each item; a user picked and placed article determining module configured to determine a user picked and placed article combination from one or more article combinations according to the attribute of each article combination; and the to-be-settled article updating module is configured to update the set of the to-be-settled articles according to the article combination taken and placed by the user.

In some embodiments, the item combination determination module is further configured to: combining one or more of the outstanding items to obtain one or more item combinations; calculating the change weight corresponding to each article combination, wherein the change weight corresponding to each article combination is equal to the weight of the goods on the shelf in the article combination minus the weight of the goods to be settled, and the weight change value is the difference between the load weight of the shelf before and after the change; and determining the article combination with the difference between the changed weight and the weight change value within a preset range as the article combination corresponding to the weight change value.

In some embodiments, the weight variation obtaining module is further configured to calculate a corresponding variation weight of each combination of items according to a pre-obtained average weight of each item; the item combination determination module is further configured to determine an item combination with a weight offset range including a weight variation value as an item combination to which the weight variation value corresponds, wherein the weight offset range is calculated from a maximum weight offset of each item in the combination.

In some embodiments, the weight change acquisition module is further configured to acquire a weight change value of a unit weighing area in response to monitoring that a load bearing weight of the unit weighing area on the shelf changes; the item combination determination module is further configured to determine one or more item combinations corresponding to weight change values from the unsettled items according to the weight of each item, wherein the unsettled items include items to be settled and items in the unit weighing area.

In some embodiments, the user picked-and-placed item determination module is further configured to determine a confidence level for each item combination according to an attribute of each item combination; selecting the article combination with the highest confidence coefficient as the article combination taken and placed by the user; wherein the attribute of each combination of items comprises at least one of: the difference between the corresponding change weight of the article combination and the change value of the weight, whether the articles in the article combination simultaneously comprise the goods on the shelf and the articles to be settled, the article type related to the article combination, the number of the article types with the number of the articles being 1 in the article combination, and whether the articles in the article combination are the articles with the largest number in the current goods channel.

In some embodiments, the article identification device further comprises: and the to-be-settled item emptying module is configured to respond to the settlement of the to-be-settled item by the user and empty the to-be-settled item.

In some embodiments, the article identification device further comprises: a category selection module configured to obtain a weight of each item of each category to be selected; performing at least one of the following screening operations: deleting the categories of which the weight is less than a preset value from the categories to be selected, and deleting at least one of two category combinations of which the weights have a multiple relation, wherein the category combination comprises one or more categories; and using the sorted goods as goods of the goods on the shelf.

In some embodiments, the article identification device further comprises: the image recognition module is configured to acquire an article image shot when a user takes and places the article on the shelf; identifying the article image to obtain an identification result; the user access item determining module is further configured to determine a user access item combination from one or more item combinations according to the attribute of each item combination and the identification result based on the image identification.

According to a third aspect of some embodiments of the present invention, there is provided an article identification apparatus comprising: a memory; and a processor coupled to the memory, the processor configured to perform any of the foregoing item identification methods based on instructions stored in the memory.

According to a fourth aspect of some embodiments of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements any one of the article identification methods described above.

Some embodiments of the above invention have the following advantages or benefits: by measuring the weight change value of the goods shelf and comparing the weight change value with a plurality of combinations of unsold goods, the goods taken and placed by the user can be accurately determined, and the accuracy of goods identification is improved. And the user can get goods, put goods, get goods and put goods simultaneously to increased the flexibility that the user used, promoted user experience. In addition, the embodiment has low manufacturing cost and is convenient to implement and deploy.

Other features of the present invention and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is an exemplary flow chart of an item identification method according to some embodiments of the invention.

Fig. 2A is an exemplary flow chart of a method of determining an item combination corresponding to a weight variation value according to some embodiments of the invention.

Fig. 2B is an exemplary flow chart of a method of determining an item combination corresponding to a weight variation value according to other embodiments of the present invention.

Fig. 3 is an exemplary flow chart of a method for identifying a combination of items accessed by a user according to some embodiments of the invention.

FIG. 4 is an exemplary flow chart of an item identification method according to further embodiments of the present invention.

Fig. 5 is an exemplary flow chart of an election method according to some embodiments of the invention.

FIG. 6 is an exemplary flow chart of an item identification method according to further embodiments of the present invention.

Fig. 7 is an exemplary flow diagram of an item identification device according to some embodiments of the invention.

Fig. 8 is an exemplary block diagram of an article identification appliance according to further embodiments of the present invention.

Fig. 9 is an exemplary block diagram of an item identification device according to further embodiments of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.

Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.

In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

FIG. 1 is an exemplary flow chart of an item identification method according to some embodiments of the invention. As shown in fig. 1, the article identification method of this embodiment includes steps S102 to S108.

In step S102, a weight change value is obtained in response to monitoring that the load weight of the shelf has changed.

The goods shelf can be the goods shelf in the automatic vending machine, and also can be the goods shelf in an unmanned supermarket, and the like. The goods shelf is provided with one or more goods channels for carrying goods. One or more weighing devices are provided in the racks for measuring the weight of the items per unit weighing area. A unit weighing area may be, for example, a lane, a row of shelves, an entire shelf, etc., which may be set as desired by one skilled in the art.

In some embodiments, when the measurement of the weighing device changes, the weighing device may send a weight change signal to inform the background, or the background may periodically and actively acquire the weight change, for example.

When the weight changes, it may be due to one of the user picking up the item from the shelf, putting the item back on the shelf, and picking up the item at the same time. When the user removes an item from the shelf, it does not necessarily mean that the user must purchase the item. The user may review the item and return it to the shelf.

In step S104, one or more item combinations corresponding to the weight change values are determined from the unsettled items according to the weight of each item.

In some embodiments, each item combination comprises at least one item, and the non-settled items comprise a shelf item and an item to be settled, wherein the shelf item is an item on the shelf before the bearing weight of the shelf changes, and the item to be settled is an item to be settled before the bearing weight of the shelf changes, namely an item which is taken from the shelf by a user and is not settled. The items to be settled can be considered as items that the user has put into a shopping cart, which can be a physical device or a virtual one. The item combination may include the source of the item, i.e. whether the item is a shelf item or an item to be settled, in addition to the name or identification and number of the item.

In some embodiments, all combinations that an outstanding item can make up may be generated. If necessary, combinations that do not significantly meet the conditions may be eliminated, and only combinations that are likely to meet the conditions may be generated. For example, for a single item having a weight greater than the weight variation value, it may be directly disregarded when generating the combination of items.

The article combination corresponding to the weight change value is the article combination of which the difference between the weight change value and the weight change value on the goods shelf caused after the goods shelf is taken and placed is within a preset range.

In step S106, an item combination picked and placed by the user is determined from one or more item combinations according to the attribute of each item combination.

When only one article combination meeting the conditions exists, the articles in the article combination can be directly determined as articles taken and placed by a user; when there are a plurality of eligible combinations of items, the combination of items with a higher possibility may be used as the combination of items taken and placed by the user by analyzing the attributes of the combination of items.

In step S108, the collection of items to be settled is updated according to the combination of the items picked and placed by the user.

For example, the shelf items in the user-picked and placed item combination are added to the collection of items to be settled, and the items to be settled in the user-picked and placed item combination are deleted from the collection of items to be settled. In addition, the set of shelf items may be updated.

According to the requirement, the process can be carried out for multiple times, namely, the user can carry out the pick-and-place process for multiple times, and the requirement of the user for multiple selection in one-time purchasing behavior is met.

In some embodiments, items to be settled may be emptied in response to a user settling the items to be settled. Such as deleting information in the shopping cart after the user pays for items in the shopping cart.

The embodiment of the invention can be applied to an application scene that a user takes and places goods and settles accounts after opening the cabinet door of the vending machine. In addition, in some application scenarios, different users can take and place goods after opening the cabinet door, but settlement is performed for the same user account during settlement. For example, after the user a performs identity authentication, the cabinet door is opened, after the cabinet door is opened, the user a and the friend B can take and place goods, and account settlement is performed by using the account of the user a during settlement.

According to the method, the weight change value of the goods shelf is measured and compared with the plurality of combinations of the unsold goods, the goods taken and placed by the user can be accurately determined, and therefore the accuracy of goods identification is improved. And the user can get goods, put goods, get goods and put goods simultaneously to increased the flexibility that the user used, promoted user experience. In addition, the embodiment has low manufacturing cost and is convenient to implement and deploy.

An embodiment of determining a combination of articles to which a weight change value corresponds is described below with reference to FIG. 2A.

Fig. 2A is an exemplary flow chart of a method of determining an item combination corresponding to a weight variation value according to some embodiments of the invention. As shown in fig. 2A, the method of this embodiment includes steps S202 to S206.

In step S202, one or more of the outstanding items are combined to obtain one or more item combinations.

The item combination comprises at least one of a shelf item and an item to be settled.

In step S204, a variation weight corresponding to each combination of items is calculated, wherein the variation weight corresponding to each combination is equal to the weight of the shelf item in the combination minus the weight of the item to be settled, and the weight variation value is the difference between the load weight of the shelf before and after variation.

Of course, the weight change value may also be the difference between the load weight of the shelf after change and the load weight before change, and the change weight corresponding to each combination is equal to the weight of the item to be settled minus the weight of the shelf item in the combination.

In step S206, the article combination with the difference between the changed weight and the weight change value within the preset range is determined as the article combination corresponding to the weight change value.

The measurement is not absolutely accurate, as the weight of the article may deviate, or the measurement may be in error. Therefore, the combination of articles having a small difference in weight from the weight change value can be determined as the eligible combination of articles.

For example, the weight of a single item of an item A on the current shelf is 210g, the weight of a single item of an item B is 250g, and if the weight of the item A is one, the weight of the item B is 2; the shopping cart has an item C weighing 380 g. All combinations of the unsettled goods are shown in table 1. The combination { A } indicates that the user removed A from the shelf, the combination { C } indicates that the user placed C to be settled back on the shelf, and the combination { B, C } indicates that the user removed B from the shelf while placing C to be settled back on the shelf.

If the weighing device on the shelf senses a weight reduction of 250g and the preset range is 250g + -5 g, { B } can be determined as the combination of items for which the weight change value corresponds.

TABLE 1

Article combination	Corresponding change weight (g)
		A	210
B	250
		C	-380
AA	420
		AB	460
AC	-170
		B C	-130
AB C	-20

The method of the above embodiment can be generally used for the identification of several situations, such as picking, putting, picking and putting at the same time. In some embodiments, the preset range may be determined from the deviation of the item itself. During the stock phase, the average weight of the items of each category may be measured as well as the maximum weight offset. An embodiment of the present invention for determining a combination of articles to which a weight change value corresponds is described below with reference to fig. 2B.

Fig. 2B is an exemplary flow chart of a method of determining an item combination corresponding to a weight variation value according to other embodiments of the present invention. As shown in fig. 2B, the method of this embodiment includes steps S212 to S216.

In step S212, one or more of the outstanding items are combined to obtain one or more item combinations.

In step S214, the variation weight corresponding to each article combination is calculated from the average weight of each article acquired in advance.

In step S216, the combination of items whose weight deviation range includes the weight variation value is determined as the combination of items corresponding to the weight variation value, wherein the weight deviation range is calculated according to the maximum weight deviation amount of each item in the combination.

By the method of the embodiment, whether the article combination meets the condition or not can be determined in an auxiliary mode according to the maximum weight offset of the articles, and the accuracy of article identification is improved.

For example, reference is still made to the foregoing examples. The weight of a single article A on a current shelf is 210g, the weight of a single article B on the current shelf is 250g, and one article A and 2 articles B are provided; the shopping cart has an item C weighing 380 g. The maximum offset of the article A, B, C was + -3 g, + -2 g, and + -4 g, respectively. All combinations of the unsettled goods are shown in table 2.

TABLE 2

Article combination	Corresponding change weight (g)	Weight deviation Range (g)
			A	210	[207,213]
B	250	[248,252]
			C	-380	[-384,-376]
AA	420	[414,426]
			AB	460	[455,465]
AC	-170	[-177,-163]
			B C	-130	[-136,-124]
AB C	-20	[-29,-11]

In some embodiments, the combination of items accessed by the user may be selected according to a confidence level of the combination of items. An embodiment of the present invention for identification of a combination of items accessed by a user is described below with reference to fig. 3.

Fig. 3 is an exemplary flow chart of a method for identifying a combination of items accessed by a user according to some embodiments of the invention. As shown in fig. 3, the method for identifying an article combination picked and placed by a user in this embodiment includes steps S302 to S304.

In step S302, a confidence level for each item combination is determined based on the attributes of each item combination.

In step S304, the item combination with the highest confidence is selected as the item combination picked and placed by the user.

Those skilled in the art can set the relationship between the attributes and the confidence levels of the combination of items as desired. Some are exemplarily described below.

Attributes that positively correlate with the confidence of the combination of items may include: the articles in the article combination are the articles with the largest quantity in the current goods channel, and the confidence coefficient is increased at the moment, so that the user is more likely to pick and place the due articles in the goods channel; the number of the article types with the article number of 1 in the article combination is higher, that is, the confidence coefficient is higher when the number of the article types with the article number of 1 is larger.

Attributes that are inversely related to the confidence of the combination of items may include: the difference between the change weight corresponding to the article combination and the weight change value is larger, and the confidence coefficient is lower; whether the items in the item combination comprise the goods shelf and the items to be settled or not at the same time is small in possibility of picking and putting the goods at the same time and low in confidence coefficient; the more the items related to the item combination, i.e. the more the number of items related to the operation of the user at one time, the lower the confidence.

For example, the weight variation value is 246g, item combination 1 includes a shelf item A, B with a corresponding variation weight of 241g, and item combination 2 includes a shelf item C and an item to be settled D with a corresponding variation weight of 245 g. The confidence of the combination of items 1 is higher if priority is given to reducing the possibility of picking and putting items at the same time; if the difference between the weight change value and the corresponding change weight of the article combination is prioritized, the confidence of the article combination 2 is higher. In addition, different types of confidence weights may be given, and the weighted sum of various confidences may be calculated as the final confidence.

The embodiment of the invention can perform partition maintenance on the articles on the shelf, thereby narrowing the selectable range of the articles to be identified. An embodiment of the article identification method of the present invention is described below with reference to fig. 4.

FIG. 4 is an exemplary flow chart of an item identification method according to further embodiments of the present invention. As shown in fig. 4, the article identification method of this embodiment includes steps S402 to S408.

In step S402, in response to monitoring that the load weight of the unit weighing area on the shelf changes, a weight change value of the unit weighing area is acquired.

A unit weighing area is the area for which a weighing device is responsible and may be, for example, a lane, or a pallet.

In step S404, one or more item combinations corresponding to the weight change value are determined from the unclassified items according to the weight of each item, wherein the unclassified items include the items to be settled and the items in the unit weighing area.

At this time, the article in the other unit weighing area is not required to be included in the article combination, and only the article in the unit weighing area and the article to be settled need to be concerned.

In step S406, an item combination picked and placed by the user is determined from the one or more item combinations according to the attribute of each item combination.

In step S408, the set of items to be settled is updated according to the combination of the items picked and placed by the user.

By the method of the embodiment, a plurality of weighing devices can be arranged in the shelf, and the combination of the articles taken and placed by the user is determined according to the articles in the unit weighing area with the changed weight, so that the accuracy of article identification is further improved.

In some embodiments, multiple data tables may be established in the database to manage each unit weighing area and shopping cart. These data tables may be updated after the combination of items accessed by the user is determined.

The invention can also optimize the selection mode aiming at the application scene of weight-based identification. An embodiment of the selection method of the present invention is described below with reference to fig. 5.

Fig. 5 is an exemplary flow chart of an election method according to some embodiments of the invention. As shown in fig. 5, the selection method of this embodiment includes steps S502 to S506.

In step S502, the weight of each item of each category to be selected is acquired.

In step S504, at least one of the following screening operations is performed: and deleting the categories of which the weight is less than the preset value from the categories to be selected, wherein at least one of the two category combinations with the weight having a multiple relation and the category combination comprise one or more categories.

When the weight of an item is too small, the weighing apparatus may have difficulty identifying the pick-and-place activity performed by the user on such an item. Screening out such items may therefore be considered. The preset value for screening can be determined according to the accuracy and sensitivity of the weighing device.

When the weight of a certain article is an integral multiple of another commodity, it is difficult to distinguish which article is taken and placed by a user. One or both may be retained. For example, if the weight of article a is 100g and the weight of article B is 200g, it is considered that only one of a and B is retained; for another example, if the weight of the article C is 200g, the weight of the article D is 170g, and the weight of the article E is 30g, only C or only D and E may be retained.

Further, confusion also readily arises when the sum of the weights of one or more items is an integer multiple of the sum of the weights of another one or more items, and so one or both sets may be retained.

Other screening methods may also be employed by those skilled in the art as desired and will not be described herein.

In step S506, the sorted item is set as the item of the shelf item.

After screening, the remaining articles are not easy to be confused in the identification process, so that the articles on the shelf can be selected from the remaining screened articles. Thus, the accuracy of article identification is further improved.

When selecting from a plurality of article combinations, the present invention can refer to the attributes of the article combinations and can perform determination by combining the image recognition results. An embodiment of the article identification method of the present invention is described below with reference to fig. 6.

FIG. 6 is an exemplary flow chart of an item identification method according to further embodiments of the present invention. As shown in fig. 6, the article identification method of this embodiment includes steps S602 to S612.

In step S602, a weight change value is obtained in response to monitoring that the load weight of the shelf has changed.

In step S604, one or more item combinations corresponding to the weight change values are determined from the unsettled items according to the weight of each item.

In step S606, an article image captured when the user picks and places an article on the shelf is acquired.

In step S608, the article image is recognized to obtain a recognition result.

For example, a camera may be disposed on or around the shelf for capturing the pick-and-place action of the user. Thus, what items the user has taken in the pick-and-place process can be displayed in the image.

In step S610, an item combination picked and placed by a user is determined from one or more item combinations according to the attribute of each item combination and the recognition result based on the image recognition.

For example, the item combination identical to the image recognition result may be added with confidence level, and the final confidence level of the item combination is calculated together with other types of confidence levels, so as to select the item combination taken and placed by the user.

In step S612, the collection of items to be settled is updated according to the combination of items picked and placed by the user.

By the method of the embodiment, gravity sensing and image recognition can be combined to more accurately recognize the articles taken and placed by the user.

An embodiment of the article identification device of the present invention is described below with reference to fig. 7.

Fig. 7 is an exemplary flow diagram of an item identification device according to some embodiments of the invention. As shown in fig. 7, the article recognition device 70 of this embodiment includes: a weight change acquisition module 710 configured to acquire a weight change value in response to monitoring that a load bearing weight of the shelf has changed; an item combination determination module 720 configured to determine one or more item combinations corresponding to the weight change values from the unsettled items according to the weight of each item; a user pick-and-place item determination module 730 configured to determine a user pick-and-place item combination from the one or more item combinations according to the attribute of each item combination; and the item to be settled update module 740 is configured to update the set of items to be settled according to the item combination taken and placed by the user.

In some embodiments, the item combination determination module 720 may be further configured to: combining one or more of the outstanding items to obtain one or more item combinations; calculating the change weight corresponding to each article combination, wherein the change weight corresponding to each article combination is equal to the weight of the goods on the shelf in the article combination minus the weight of the goods to be settled, and the weight change value is the difference between the load weight of the shelf before and after the change; and determining the article combination with the difference between the changed weight and the weight change value within a preset range as the article combination corresponding to the weight change value.

In some embodiments, the weight variation acquisition module 710 may be further configured to calculate a corresponding variation weight for each combination from the pre-acquired average weight for each item; the item combination determination module is further configured to determine a combination of weight deviation ranges including the weight variation value as the item combination to which the weight variation value corresponds, wherein the weight deviation ranges are calculated according to the maximum weight deviation amount of each item in the combination.

In some embodiments, the weight change acquisition module 710 may be further configured to acquire a weight change value for a unit weighing area on the shelf in response to monitoring that the load bearing weight of the unit weighing area changes; the item combination determination module 720 may be further configured to determine one or more item combinations corresponding to the weight variation values from the outstanding items according to the weight of each item, wherein the outstanding items include the items to be settled, and the items in the unit weighing area.

In some embodiments, the user picked-and-placed item determination module 730 may be further configured to determine a confidence level for each item combination according to the attributes of each item combination; selecting the article combination with the highest confidence coefficient as the article combination taken and placed by the user; wherein the attribute of each combination of items comprises at least one of: the difference between the corresponding change weight of the article combination and the change value of the weight, whether the articles in the article combination simultaneously comprise the goods on the shelf and the articles to be settled, the article type related to the article combination, the number of the article types with the number of the articles being 1 in the article combination, and whether the articles in the article combination are the articles with the largest number in the current goods channel.

In some embodiments, the article identification device 70 may further include: and an item to be settled emptying module 750 configured to empty the item to be settled in response to the user settling the item to be settled.

In some embodiments, the article identification device 70 may further include: a category selection module 760 configured to obtain a weight of each item of each category to be selected; performing at least one of the following screening operations: deleting the categories of which the weight is less than a preset value from the categories to be selected, and deleting at least one of two category combinations of which the weights have a multiple relation, wherein the category combination comprises one or more categories; and using the sorted goods as goods of the goods on the shelf.

In some embodiments, the article identification device 70 may further include: an image recognition module 770 configured to acquire an item image taken by a user when picking and placing an item on a shelf; identifying the article image to obtain an identification result; the user access item determination module 730 may be further configured to determine a user access item combination from one or more item combinations according to the attribute of each item combination and the recognition result based on the image recognition.

Fig. 8 is an exemplary block diagram of an article identification appliance according to further embodiments of the present invention. As shown in fig. 8, the article recognition apparatus 800 of this embodiment includes: a memory 810 and a processor 820 coupled to the memory 810, the processor 820 being configured to perform the method of item identification in any of the preceding embodiments based on instructions stored in the memory 810.

Memory 810 may include, for example, system memory, fixed non-volatile storage media, and the like. The system memory stores, for example, an operating system, an application program, a Boot Loader (Boot Loader), and other programs.

Fig. 9 is an exemplary block diagram of an item identification device according to further embodiments of the present invention. As shown in fig. 9, the article recognition apparatus 900 of this embodiment includes: the memory 910 and the processor 920 may further include an input/output interface 930, a network interface 940, a storage interface 950, and the like. These

interfaces

930, 940, 950 and the memory 910 and the processor 920 may be connected, for example, by a bus 960. The input/output interface 930 provides a connection interface for input/output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 940 provides a connection interface for various networking devices. The storage interface 950 provides a connection interface for external storage devices such as an SD card and a usb disk.

An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, wherein the program is configured to implement any one of the article identification methods described above when executed by a processor.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims

1. An item identification method comprising:

acquiring a weight change value in response to monitoring that the load bearing weight of the goods shelf changes;

determining one or more item combinations corresponding to the weight change value from the non-settled items according to the weight of each item, wherein each item combination comprises at least one item, the non-settled items comprise shelf items and items to be settled, the shelf items are the items on the shelf before the bearing weight of the shelf changes, the items to be settled are the items to be settled before the bearing weight of the shelf changes, and the determining one or more item combinations corresponding to the weight change value from the non-settled items comprises: combining one or more of the outstanding items to obtain one or more item combinations; calculating the change weight corresponding to each item combination, wherein the change weight corresponding to each item combination is equal to the weight of the goods on the shelf in the item combination minus the weight of the goods to be settled, and the weight change value is the difference between the load weight of the shelf before and after the change; determining the article combination with the difference between the changed weight and the weight change value within a preset range as the article combination corresponding to the weight change value;

determining an article combination taken and placed by a user from the one or more article combinations according to the attribute of each article combination;

and updating the set of the articles to be settled according to the article combination taken and placed by the user.

2. The item identification method according to claim 1,

calculating the corresponding change weight of each article combination according to the average weight of each article obtained in advance;

and determining the article combination with the weight deviation range including the weight change value as the article combination corresponding to the weight change value, wherein the weight deviation range is calculated according to the maximum weight deviation amount of each article in the combination.

3. The item identification method according to claim 1,

in response to monitoring that the bearing weight of a unit weighing area on a shelf changes, acquiring a weight change value of the unit weighing area;

and determining one or more article combinations corresponding to the weight change value from the unsettled articles according to the weight of each article, wherein the unsettled articles comprise articles to be settled and articles in the unit weighing area.

4. The item identification method according to claim 1,

determining a confidence level of each item combination according to the attribute of each item combination;

selecting the article combination with the highest confidence coefficient as the article combination taken and placed by the user;

wherein the attribute of each combination of items comprises at least one of: the difference between the change weight corresponding to the article combination and the weight change value, whether the articles in the article combination simultaneously comprise the goods on the shelf and the articles to be settled, the article type related to the article combination, the number of the article types with the number of the articles being 1 in the article combination, and whether the articles in the article combination are the articles with the largest number in the current goods channel.

5. The item identification method of claim 1, further comprising:

and responding to the settlement of the article to be settled by the user, and emptying the article to be settled.

6. The item identification method of claim 1, further comprising:

acquiring the weight of each article of each to-be-selected category;

performing at least one of the following screening operations: deleting a category of which the weight is less than a preset value from the categories to be selected, and deleting at least one of two category combinations of which the weights have a multiple relation, wherein the category combinations comprise one or more categories;

and using the sorted goods as goods of the goods on the shelf.

7. The article identification method according to claim 1,

further comprising: acquiring an article image shot when a user takes and places articles on a shelf; identifying the article image to obtain an identification result;

the determining, according to the attribute of each item combination, an item combination picked and placed by a user from the one or more item combinations comprises: and determining the article combination taken and placed by the user from the one or more article combinations according to the attribute of each article combination and the identification result based on the image identification.

8. An article identification device comprising:

the weight change acquisition module is configured to respond to the monitored change of the bearing weight of the goods shelf and acquire a weight change value;

an item combination determination module configured to determine, according to a weight of each item, one or more item combinations corresponding to the weight change value from the outstanding items, where each item combination includes at least one item, the outstanding items include a shelf item and an item to be settled, the shelf item is an item located on the shelf before a change in a bearing weight of the shelf occurs, and the item to be settled is an item to be settled before the change in the bearing weight of the shelf occurs, and the determining, from the outstanding items, the one or more item combinations corresponding to the weight change value includes: combining one or more of the outstanding items to obtain one or more item combinations; calculating the change weight corresponding to each item combination, wherein the change weight corresponding to each item combination is equal to the weight of the goods on the shelf in the item combination minus the weight of the goods to be settled, and the weight change value is the difference between the load weight of the shelf before and after the change; determining the article combination with the difference between the changed weight and the weight change value within a preset range as the article combination corresponding to the weight change value;

a user picked and placed article determining module configured to determine a user picked and placed article combination from the one or more article combinations according to the attribute of each article combination;

and the to-be-settled item updating module is configured to update the set of to-be-settled items according to the item combination taken and placed by the user.

9. The article identification device of claim 8,

the weight change acquisition module is further configured to calculate a corresponding change weight of each combination of items according to a pre-acquired average weight of each item;

the item combination determination module is further configured to determine an item combination with a weight offset range including the weight variation value as an item combination to which the weight variation value corresponds, wherein the weight offset range is calculated according to a maximum weight offset of each item in the combination.

10. The article identification device of claim 8,

the weight change acquisition module is further configured to acquire a weight change value of a unit weighing area on a shelf in response to monitoring that the load bearing weight of the unit weighing area changes;

the item combination determination module is further configured to determine one or more item combinations corresponding to the weight change values from the unsettled items according to the weight of each item, wherein the unsettled items include items to be settled and items in the unit weighing area.

11. The item identification device of claim 8, wherein the user picked and placed item determination module is further configured to determine a confidence level for each item combination based on an attribute of each item combination; selecting the article combination with the highest confidence coefficient as the article combination taken and placed by the user;

12. The article identification device of claim 8, further comprising:

and the clearing module of the article to be settled is configured to respond to the user clearing the article to be settled and clear the article to be settled.

13. The article identification device of claim 8, further comprising:

a category selection module configured to obtain a weight of each item of each category to be selected; performing at least one of the following screening operations: deleting a category of which the weight is less than a preset value from the categories to be selected, and deleting at least one of two category combinations of which the weights have a multiple relation, wherein the category combinations comprise one or more categories; and using the sorted goods as goods of the goods on the shelf.

14. The article identification device as claimed in claim 8,

further comprising: the image recognition module is configured to acquire an article image shot when a user takes and places the article on the shelf; identifying the article image to obtain an identification result;

the user access item determination module is further configured to determine a user access item combination from the one or more item combinations according to the attribute of each item combination and the identification result based on the image recognition.

15. An article identification device comprising:

a memory; and

a processor coupled to the memory, the processor configured to perform the item identification method of any of claims 1-7 based on instructions stored in the memory.

16. A computer-readable storage medium, on which a computer program is stored, which program, when executed by a processor, implements the item identification method according to any one of claims 1 to 7.